Simple Laboratory and Clinical Parameters as Predictor for Electrocardiographic Abnormalities Among Hospitalized Patients with Chronic Kidney Disease: A Cross-Sectional Study

Abstract

BackgroundIn developing countries, even electrocardiography (ECG) hasn’t been used widely in most health-care centers. The ability of physicians to refer to chronic kidney disease (CKD) patients for ECG, often collide with several barriers and costs. Therefore, we need to formulate the simplest and most efficient model to predict when CKD patients need to be referred due to potential ECG abnormalities.ObjectiveThe aim of this study was to develop several clinical and laboratory parameters as a predictor of any ECG abnormalities.Materials and MethodsA retrospective cross-sectional study design held at Dr. Soetomo General Academic Hospital, Surabaya, Indonesia. Subjects were hospitalized patients with CKD between 1 January to 31 December 2019. 198 CKD patients (101 males) were enrolled for the study. All patients had demographic information, detailed clinical profile, resting 12-lead ECG recording, complete blood count, serum electrolyte and renal function test profile during admission and results were interpreted blindly by two cardiologists. Statistical analysis was done by SPSS 17.0.ResultsA total of 198 patients were included in this study. Mean ages were 52.2±11.8 years old and fifty-one percent were males. Eighty-eight percent of patients from 198 patients had ECG abnormality. AUC of hemoglobin level to discriminate poor R wave progression, pathological Q wave, non-spesific ST-T changes, and frontal axis deviation were 0.532, 0.641, 0.556 and 0.693, respectively. In multivariate logistic regression analysis, only higher systolic blood pressure was determined as an independent predictor of abnormal ECG finding in CKD patients, as systolic blood pressure increase by one unit, the odds of having abnormal ECG is increased 1.02 times (95% CI: 1.00 – 1.02, p=0.042).ConclusionThe ECG abnormalities can be found in hospitalized CKD patients. Fragmented QRS and long QTc were the highest prevalent ECG abnormalities in our study. Serum creatinine and hemoglobin could predict peaked T wave and prolonged QTc among hospitalized CKD patients. Systolic blood pressure could predict prolonged QTc and fragmented QRS in CKD patients.